The Agent Skills Directory

[SAFE]: No security issues or malicious patterns were identified in the skill instructions or metadata.
[DATA_EXPOSURE]: The skill reads the memory/user-profile.md file to provide context-aware suggestions. This is a localized read for personalization and does not involve transmitting sensitive data to external sources.
[INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface through user-provided arguments and web search results. However, its capabilities are restricted to generating structured text output and writing to the designated workspace/ directory, which is consistent with its intended brainstorming purpose.
Ingestion points: $ARGUMENTS and WebSearch tool output
Boundary markers: Absent
Capability inventory: Read and Write tool access; writes to workspace/brainstorm-[slug].md (file-write)
Sanitization: Absent

brainstorm